Policy Research Working Paper 11257 What Underlies the Poor Financial Performance of Electric Utilities in Sub-Saharan Africa? Govinda R Timilsina Development Economics Development Research Group November 2025 Policy Research Working Paper 11257 Abstract This study investigates the factors responsible for the poor consumers face relatively higher electricity tariffs than in performance of 67 electric utilities in 47 countries, using many countries around the world. The study also finds that descriptive data from the World Bank, the International if the transmission and distribution losses were reduced Energy Agency, the U.S. Energy Information Administra- to the current level of South Africa (11 percent) and the tion, and national sources. The findings show that both leakages in bill collection were eliminated, several electric cost-side and revenue-side factors are responsible for the utilities that are currently operating at a loss would have poor financial performance of electric utilities. More than higher revenue than their operational cost. The findings two-thirds of vertically integrated utilities and electricity indicate that policy makers in the region should focus on distribution utilities are unable to cover their operational a portfolio of policies, including switching from expensive and debt service costs by their revenues. The main causes generation to emerging cheaper options, improving factor of the poor financial performance are high fuel costs (par- productivities, having efficient institutions and governance, ticularly oil), low capacity factors, low capital and labor reducing transmission and distribution losses, improving productivity, high transmission and distribution losses, and bill collection, and reforming tariffs. The policy priorities leakage in electricity bill collections. The study finds that in could vary across countries, depending on the roles of var- these countries, despite their much lower per capita income, ious factors contributing to poor financial performance. This paper is a product of the Development Research Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The author may be contacted at gtimilsina@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team 1. Introduction The Sub-Saharan Africa (SSA) region faces a significant challenge to increase access to electricity. Almost 600 million people (or 43% of the total population) in the SSA region do not have access to electricity (World Bank, 2025a). The situation is worse in the rural areas, with only 26% of the population having access to electricity (World Bank, 2025a). Even if there is access, the quality of supply is poor due to hours of scheduled outages of electricity (loadshedding) every day, frequent unscheduled outages, and voltage drops and fluctuations (Hafner et al. 2018). One of the primary reasons for lack of access and low quality of supply is the inability of electricity utilities to efficiently operate the existing supply capacity and to add new capacity. This inability is linked with poor financial performance characterized by much higher operational and debt service costs as compared to their revenue, despite the direct government subsidies they receive for a prolonged time. Institutional inefficiency is the major factor behind the high operational costs per unit of electricity generation and lower revenue generation per unit of electricity sold. This study evaluates the performance of 67 electricity utilities in the SSA region for which data is available. 1 These utilities are divided into six groups based on their functional responsibilities. These are (i) vertically integrated utilities (VIU) which provide all three functions (i.e., electricity generation, transmission and distribution) for supplying electricity services; (ii) electricity generation utilities (EGUs) which generate electricity and sell to electricity transmission and/or distribution utilities/companies; (iii) electricity generation and transmission utilities (EGTs) which are responsible for electricity generation and transmission, and sell electricity to electricity distributors (e.g., distribution utilities, private companies, community groups, municipalities); (iv) electric utilities for transmission purpose only (ETUs), these utilities buys electricity from generators and sale to distributors; (v) electricity utilities with both transmission and distribution (T&D) responsibilities (TDUs) which buy electricity from EGUs and perform the transmission and distribution services; and finally (vi) electricity distribution utilities (EDUs) whose function is to distribute electricity to 1 See Table A1 in the Appendix A for the list of these utilities. 2 consumers buying it from EGUs or EGTs or ETUs. Of the 67 utilities considered, 30 are VIUs, eight are EGUs, two are EGTs, seven are ETU, 19 are EDU and three are TDUs. All VIUs are publicly owned except Cameron’s ENEO. ETUs and TDUs are, by virtue, publicly owned. While all EDUs in Nigeria and an EDU in Uganda are privately owned, the rest of the EDUs in SSA are publicly owned. The size of the utilities varies significantly in terms of their assets and employment, the number of customers served, and the electricity generated by VIUs and EGUs or sold by VIUs, EDUs, and TDUs. Figure A1 presents a comparison of utilities in terms of these size indicators. In terms of total assets of the VIUs, South Africa’s ESKOM is the largest utility in the SSA region, followed by the Democratic Republic of Congo’s SNEL, Tanzania’s TANSCO and Côte d'Ivoire’s CIENERGIES. The total assets of EGUs are also higher as they own dams and generation machineries. 2 This is also true for ETUs and TDUs because they own transmission towers, transmission wires, switchyards with large transformers. EDUs normally have lower assets because they own distribution transformers and distribution wires only. In terms of employment, the ranking of the SSA electric utilities is similar to that based on asset values (see Figure A1b in the Appendix A). Some existing studies analyze the financial performance data of electric utilities in the SSA region (e.g., Balabanyan et al. 2021; Trimble et al. 2016; Kozima and Trimble, 2016). Using publicly available financial and operational performance data of 76 electric utilities in 45 countries in the SSA region for the 2012-2018 period, Balabanyan et al. (2021) report several findings. First, the cost recovery by revenue worsened between 2012 and 2018 despite the 16% increase in electricity tariff. Second, utilities face growing liquidity problems due to poor electricity bill collection rates and longer accounts receivable periods. Third, electricity distribution loss increased rather than decreased over time. Earlier, Kozima and Trimble (2016) analyze the technical and financial performances of electric utilities in 39 SSA countries for three years, 2012 to 2014. They also report that most electric utilities in the region struggle financially due to aging infrastructure and 2 If the generation, transmission, and distribution assets are added, Ethiopia is the second largest country after South Africa in the SSA region in terms of electric utility assets. 3 inefficiencies. They suggest minimizing T&D losses, improving the bill collection rate and increasing tariffs for larger consumers to improve the financial performance of utilities. Utilizing the same data set used in Kozima and Trimble (2016), Trimble et al. (2016) calculate quasi-fiscal deficits of electric utilities and conclude that electric utilities will not be able to maintain existing assets let alone expand the capacity for universal access. An earlier study (Fritsch, 2011) compares some financial indicators of electric utilities in West African countries and argues that increasing the earnings before interest, taxes, debt and amortization (EBITDA) through increased tariff, reduction of technical losses and diversification of oil and gas-based generation is necessary to improve the financial performance of these utilities. Our study contributes to the literature in several ways. First, it uses the most recent data as compared to existing studies which use data that are at least 10 years old, and captures the changes in financial performance since then. Second, it examines additional factors responsible for poor financial performance of electric utilities in the SSA region, such as higher fuel costs, lower capacity factors and lower capital and labor productivities. Third, it calculates the amount of revenue loss that can be avoided through reduction of T&D loss and improving the electricity bill collection rates. Fourth, it highlights the role of predominant use of expensive fuels (i.e., petroleum products) with inefficient generation technologies as one of the main contributors to the high costs of electricity production in the region. This factor is ignored in existing literature. The study finds both cost-side and revenue-side factors responsible for the poor financial performance of electric utilities in the SSA region. Higher T&D loss and lower collection rates are mainly responsible for revenue loss. Higher fuel costs, lower thermal efficiency and capacity utilization rates and lower labor and capital productivities are responsible for higher operational costs. While electricity tariffs are often blamed for the poor financial performance of electric utilities, arguing that tariffs are set lower than production costs (e.g., Burgess et al. 2020), our study finds that electricity tariffs of SSA utilities are higher than those of other utilities around the world except those of OECD utilities. Based on these findings, our study suggests that policy measures should be focused on technical factors on the cost side (e.g., reducing production 4 costs through fuel switching and increasing factor productivities) as well as the revenue side (e.g., reducing T&D loss and bill collection leakage) instead of increasing electricity tariffs in low-income SSA countries. However, existing subsidies either provided to electricity directly or through fossil fuels used for electricity should be reduced. While doing so cross- subsidization of electricity through rate design to make it affordable (or provision of life-line tariff) for the poor is a normal practice around the world. The rest of the paper is organized as follows. Section 2 evaluates the financial performance of electric utilities, followed by discussion of key factors contributing to the financial gaps in Section 3. Section 4 estimates the improvements in financial performance through T&D loss reduction and elimination of collection leakage and highlights operational cost cutting measures. Key conclusions are drawn in Section 5. 2. Financial performance of electric utilities One of the biggest challenges that SSA electric utilities are currently facing is their financial performance. A majority of utilities have annual revenues lower than their operational costs. If the cost of debt service 3 is also accounted for to reflect the cost of capital, even more utilities are operating at a loss because they are unable to generate enough revenues to cover operational and debt service cost. Figure 1 (left panel) presents the ratios of average operational costs to average revenue and average operational plus debt cost to average revenue. Of 30 VIUs, 15 have their average operational costs higher than their average revenue. If the cost of debt is also included, another five VIUs are in loss. In some cases, operational costs are more than double their revenues. The financial performance of EDUs is not better than that of VIUs despite the fact that overall operational costs of the former are much smaller than those of the latter. Of the 19 EDUs we have considered, only five EDUs (Ethiopia’s EEU, Namibia’s ERONGO and CENORED, Nigeria’s IKEJA and Uganda’s UMEME) have revenues higher than their 3 The debt service cost, which includes the interest expense and repayment component of debt servicing cashflows, reflects the capital cost. 5 operational and debt service costs (see Figure 1, right panel). As expected, ETUs have revenues higher than operational costs, as the electricity transmission system has relatively lower operational costs. This is also the case for EGUs, particularly those with predominantly hydropower assets. In the case of Ethiopia’s EEP, the cost of debt service is three times higher than its revenue because of large-scale hydropower plants more recently commissioned. It will take several years to pay back the loans taken by this utility and turn it into a profit-making one. Figure 1. Ratios of average operational costs and average operational plus debt service costs to average revenue of electric utilities in the SSA region Zimbawe (ZESCO) Togo (CEET) Zimbabwe (ZPC) Tanzania (TANESCO) Uganda (UEGCL) Sao Tome & Principe… EGU Kenya (KenGen) South Africa (ESKOM) Seychelles (PUC) Ghana (VRA) Senegal (SENELEC) Angola (PRODEL) Ruwanda (EUCL) TDU EGT Ethiopia (EEP) Niger (NIGELEC) Namibia (NAMPOWER) Kenya (KPLC) Mozambique (EDM-… Cote d' Ivoire (CIE) Mauritius (CEB) Mali (EDM) Uganda (UETCL) ETU Malawi (ESCOM) Angola (RNT) VIU (Vertically Integrated Utilities) Madagascar (JIRMA) Uganda (UMEME) Liberia (LECLIB) Lesotho (LECLES) Nigeria (YEDC) Guinea (EDG) Nigeria (PHED) Gambia (NAWEC) Nigeria (KEDCO) Gabon (SEEG) Cote d' Ivoire… Nigeria (IKEJA) DR Congo (SNEL) Nigeria (EKEDC) EDU Chad (SNE) Nigeria (AEDC) Central African Rep.… Cameroon (ENEO) Namibia (ERONGO) Cabo Verde (ELECTRA) Namibia (CENORED) Burundi (REGIDESO) Ghana (ECG) Burkina Faso (SONABEL) Botswana (BPC) Ethiopia (EEU) Benin (SEBE) Angola (ENDE) 0.00 1.00 2.00 3.00 4.00 5.00 0.0 2.0 4.0 Operational and debt service cost to revenue ratio Operational cost to revenue ratio Operational cost to revenue ratio Operational and debt service cost to revenue ratio Source: World Bank (2025c) 6 Most governments in SSA have provided direct subsidies to electric utilities to reduce their stress due to poor financial performance and to keep electricity prices lower for consumers. The subsidy helps utilities to have higher revenue than their operational and debt service costs because subsidies are cash inflows to the utilities. However, the financial gaps (i.e., operational and debt service costs in excess of revenues) are too large in those utilities which are already in financial loss (i.e., utilities which have lower revenues than their operational and debt service costs). Only eight utilities of the 40 (including all types of utilities) that are in financial loss can cover their operational and debt service costs (Figure 2) when subsidy is accounted for. Specifically, five VIUs and three other utilities do so. The situation has not changed much for many electric utilities in the region over the years. An earlier study, Fritsch (2011) shows that none of seven electric utilities in West Africa (CIE, CEET, EDM, NIGELEC, SBEE, SENELEC and SONABEL) generated enough revenue to cover their operational plus debt service costs even if the subsidy provided for their operation was accounted for in 2009. These utilities, except CIE which provides transmission service only, and SBEE are still in financial loss more recently, after 2020. Analyzing the financial performance of 76 electric utilities in 45 countries in the SSA region, Balabanyan et al. (2021) also report that more than 60% of the utilities did not recover their operational and debt-service costs with their revenue in 2018. Eight of the 27 utilities that recovered operational and debt service costs could not do so without operational subsidies. Using 2012-2014 utility data, Trimble (2016) showed only two countries in SSA had financially viable electricity sectors (the Seychelles and Uganda) during that period. 7 50 100 150 200 0 100 150 0 50 Angola (ENDE) Benin (SEBE) Botswana (BPC) Ethiopia (EEU) Burkina Faso… Burundi (REGIDESO) Ghana (ECG) Cabo Verde (ELECTRA) Cen. Afr. Rep. (ENERCA) Namibia (CENORED) Chad (SNE) Cote d' Ivoire… Source: World Bank (2025c) Namibia (ERONGO) Gabon (SEEG) Namibia (NORED) Gambia (NAWEC) VIU Guinea (EDG) Nigeria (AEDC) Lesotho (LECLES) Liberia (LECLIB) Nigeria (BEDC) Madagascar (JIRMA) Excluding subsidy 8 Malawi (ESCOM) Nigeria (EKEDC) Mali (EDM) Excluding subsidies subsidy (%) Nigeria (IBEDC) Mauritania (SOMELEC) Mauritius (CEB) Nigeria (IKEJA) Mozambique (EDMMOZ) Namibia (NAMPOWER) Including subsidy inefficiency factors, such as low labor and capital productivity. Nigeria (JOS) Niger (NIGELEC) Ruwanda (EUCL) Including subsidies Nigeria (KEDCO) Senegal (SENELEC) Seychelles (PUC) Nigeria (PHED) 3. Factors behind the poor financial performance of utilities South Africa (ESKOM) Nigeria (YEDC) Sao Tome & Principe… Tanzania (TANESCO) Sudan (SEDC) Togo (CEET) Zimbawe (ZESCO) Uganda (UMEME) Cameroon (ENEO) Figure 2. Coverage of operational and debt service costs by revenues with and without Why is the financial performance of electric utilities so poor in SSA? We investigate this question by analyzing some of the key factors including existing electricity tariffs, revenue losses due to T&D losses, missing electricity bill collections and other supply-side 3.1 Cost factors 3.1.1 Higher operational costs One of the key reasons for the poor financial performance of electric utilities in SSA is their very high operational costs. The operational costs of VIUs in the region vary from US$70 to US$671per MWh, with median value of US$183/MWh (see upper panel of Figure 3). We compared these costs with operational costs of utilities in other regions where the utilities also have vertically integrated structures. These are electric utilities in Saudi Arabia (Middle East & North Africa region), Nepal (South Asia region) and Viet Nam (East Asia & Pacific region). The operational costs in Saudi Arabia, Nepal and Viet Nam are respectively, US$42, US$63 and US$72 per MWh (World Bank, 2025c). These numbers are much smaller as compared to the median operational costs of VIUs in the SSA region. The operational costs of electricity distribution utilities (e.g., EDUs) are presented in the lower panel of Figure 3. We have not presented here operational costs of utilities that provide either transmission or generation operations because their operational costs are not comparable with that of EDUs due to difference in functional structures. Operational costs vary significantly across EDUs in the SSA region. While operational costs of EDUs in Angola and Ethiopia are less than US$50/MWh, operational costs of most of the remaining EDUs exceed US$150/MWh. We compared these costs with some EDUs in other countries in different regions: Empresa Distribuidora de Electricidad de Mendoza S.A. (EDEM) of Argentina, Bangalore Electricity Supply Company Limited (BESCL) of India and Metropolitan Electricity Authority (MEA) of Thailand. The operational costs of EDEM, BESCL and MEA are, respectively, US$78, US$103 and US$113 per MWh. The median value of operational costs of these EDUs is US$154/MWh. There are two interesting findings here. First, the difference between the median values of operational costs of VIUs and EDUs are not very different (US$183 versus US$154 per MWh) because most of the operational costs, particularly in the case of fossil fuel dominant power systems, occur at the generation segment as compared to transmission and distribution segments. Second, the median value of operational costs 9 of EDUs in SSA is closer to that of EDUs in the different regions outside SSA considered here, although some EDUs in SSA have operational costs that are more than double those of the utilities outside SSA used for comparison purposes here. Why are the average operational costs so high for SSA electric utilities? One of the key components of the total operational cost is electricity generation cost, which includes cost of fuels used for electricity generation and costs for regular maintenance of power plants, transmission and distribution systems, cost of labor (i.e., wage expenses), and cost of capital (i.e., cost of debt) if we also account for debt service costs. We will discuss the level of these costs across the utilities as far as data is available. Figure 3. Average operational costs including and excluding debt service costs (US$/MWh) 600 Average operational cost Average debt service costs 500 400 300 200 100 0 Benin (SEBE) Congo, Dem. Rep. (SNEL) Burkina Faso (SONABEL) Malawi (ESCOM) Ruwanda (EUCL) Burundi (REGIDESO) Chad (SNE) Cote d' Ivoire (CIENERGIES) Liberia (LECLIB) Madagascar (JIRMA) Namibia (NAMPOWER) Niger (NIGELEC) Seychelles (PUC) Sao Tome & Principe (EMAE) Togo (CEET) Zimbawe (ZESCO) Cameroon (ENEO) Botswana (BPC) Cabo Verde (ELECTRA) Cen. Afr. Rep. (ENERCA) Gabon (SEEG) Guinea (EDG) Mali (EDM) Mauritius (CEB) Senegal (SENELEC) Gambia (NAWEC) Lesotho (LECLES) Tanzania (TANESCO) Mozambique (EDM-MOZ) South Africa (ESKOM) VIU 10 250 Average operational cost 200 Average debt service cost 150 100 50 0 Namibia (CENORED) Namibia (ERONGO) Nigeria (PHED) Nigeria (YEDC) Ghana (VRA) Kenya (KenGen) Nigeria (KEDCO) Cote d' Ivoire (CIE) Ethiopia (EEP) Ghana (ECG) Nigeria (AEDC) Angola (PRODEL) Uganda (UEGCL) Angola (ENDE) Ethiopia (EEU) Nigeria (EKEDC) Nigeria (IKEJA) Angola (RNT) Uganda (UMEME) Kenya (KPLC) Zimbabwe (ZPC) Uganda (UETCL) EDU ETU TDU EGT EGU Source: World Bank (2025c) 3.1.2 Fuel costs The data on fuel costs are often not available, and even when they are available, it is difficult to distinguish whether the fuel, particularly diesel, is used in power plants or transportation vehicles. Data on generation mix also helps to explain the role of fuel costs to increase electricity production costs. Table 1 presents generation mix data for 2023. Of the 43 countries in Table 1, 17 countries produce more than half of their total generation through hydropower plants, and 13 countries do the same through oil-fired power plants. Natural gas is the predominant source of power generation in 7 countries and coal in two countries. Oil, particularly diesel, is the most expensive fuel for power generation and it is mainly imported in SSA countries. Moreover, diesel is used in internal combustion engines which have the least thermal efficiency. Due to the high fuel prices and low thermal efficiency, electric utilities where oil products, particularly diesel, are the predominant source of electricity, face higher operational costs. Figure 4 plots operational costs vis-a-vis share of oil-based generation for VIUs in the SSA region. The figure shows that electric utilities where oil is the predominant source of generation have higher operational costs (e.g., Chad, Cabo Verde, 11 Mali, São Tomé and Príncipe, Senegal). 4 On the other hand, electric utilities where hydro or natural gas is the predominant source of generation have relatively lower operational costs (e.g., Angola, Benin, Cameron, Côte d’Ivoire, the Democratic Republic of Congo, Ethiopia, Mozambique, Namibia, Tanzania, Zimbabwe). This is because hydropower does not have fuel cost, and the price of domestically produced natural gas is low in SSA countries due to the lack of export infrastructure. Figure 4. Share of generation in total generation vis-à-vis operational costs 100 700 90 % in total generation Operation costs (US$/MWh) 600 80 70 500 60 400 50 40 300 30 200 20 10 100 0 0 Cen. Afr. Rep.… Sao Tome & Principe… Cote d' Ivoire… Congo, Dem. Rep.… Burkina Faso… Mozambique… Benin (SEBE) Ghana (VRA) Ruwanda (EUCL) Burundi (REGIDESO) Cameroon (ENEO) Malawi (ESCOM) Chad (SNE) Ethiopia (EEP) Liberia (LECLIB) Namibia (NAMPOWER) Niger (NIGELEC) Seychelles (PUC) Togo (CEET) Zimbawe (ZESCO) Angola (PRODEL) Botswana (BPC) Cabo Verde (ELECTRA) Guinea (EDG) Senegal (SENELEC) Gambia (NAWEC) Mali (EDM) Lesotho (LECLES) Tanzania (TANESCO) South Africa (ESKOM) Coal Oil Natural Gas Hydro Others Operational costs Note: Others include nuclear, geothermal, solar, wind and biomass 4 Oil-based power plants are relatively inefficient with 28% thermal efficiency of diesel-fired internal combustion engine and 33% thermal efficiency of fuel oil fired steam turbine. According to Statistica (https://www.statista.com/statistics/1297071/average-retail-prices-for-diesel-in-africa-by-country), diesel prices in the SSA region varied from US$1.26 in Senegal to US$2.24 in the Central African Republic in 2023. With 28% efficiency, US$2.24 per liter is equivalent to US$752/MWh fuel cost in power plants. 12 Table 1. Electricity generation mix in 2023 (GWh and %) Total % of the total generation generation Natural Country (GWh) Coal Oil Gas Hydro Nuclear Solar Wind Geothermal Biomass Angola 17,939 0.0 14.3 9.3 74.0 0.0 2.2 0.0 0.0 0.3 Benin 1,001 0.0 23.6 73.1 0.0 0.0 3.3 0.0 0.0 0.0 Botswana 2,584 95.9 3.9 0.0 0.0 0.0 0.2 0.0 0.0 0.0 Burkina Faso 1,732 0.0 82.9 0.0 6.7 0.0 5.4 0.0 0.0 5.1 Burundi 384 0.0 31.2 0.0 66.7 0.0 0.5 0.0 0.0 1.6 Cabo Verde 506 0.0 71.1 0.0 0.0 0.0 14.2 14.6 0.0 0.0 Cameroon 8,339 0.0 12.6 23.5 63.1 0.0 0.3 0.0 0.0 0.5 Central African Republic 142 0.0 0.8 0.0 99.2 0.0 0.0 0.0 0.0 0.0 Chad 391 0.0 94.3 0.0 0.0 0.0 0.8 2.3 0.0 2.6 Côte d'Ivoire 11,132 0.0 0.2 68.7 30.1 0.0 0.2 0.0 0.0 0.8 Congo, Dem. Rep. 15,900 0.0 0.0 0.0 86.0 0.0 13.8 0.0 0.0 0.2 Ethiopia 18,254 0.0 0.0 0.0 96.5 0.0 0.2 3.1 0.2 0.0 Gabon 3,193 0.0 16.9 35.0 47.7 0.0 0.0 0.0 0.0 0.3 Gambia, The 515 0.0 99.0 0.0 0.0 0.0 0.6 0.4 0.0 0.0 Ghana 24,282 0.0 1.8 59.7 37.8 0.0 0.6 0.0 0.0 0.1 Guinea 4,048 0.0 25.3 0.0 74.1 0.0 0.6 0.0 0.0 0.0 Kenya 12,789 0.0 10.2 0.0 20.9 0.0 4.5 15.7 47.2 1.6 Lesotho 482 0.0 0.1 0.0 99.6 0.0 0.3 0.0 0.0 0.0 Liberia 395 0.0 66.1 0.0 32.4 0.0 1.3 0.0 0.0 0.3 Madagascar 2,645 19.3 45.5 0.0 31.1 0.0 3.2 0.0 0.0 0.9 Malawi 1,837 0.0 4.2 0.0 92.2 0.0 0.7 0.0 0.0 3.0 Mali 4,363 0.0 57.3 0.0 37.6 0.0 3.5 0.0 0.0 1.6 Mauritania 1,642 0.0 72.4 0.0 12.8 0.0 8.5 6.3 0.0 0.0 Mauritius 3,264 33.5 48.9 0.0 2.9 0.0 4.6 0.3 0.0 9.9 Mozambique 19,559 0.0 0.6 15.7 82.7 0.0 0.4 0.0 0.0 0.6 Namibia 1,890 1.6 0.3 0.0 70.0 0.0 26.9 1.2 0.0 0.0 Niger 804 22.4 68.4 6.2 0.0 0.0 3.0 0.0 0.0 0.0 13 Nigeria 42,509 0.0 0.0 77.1 22.5 0.0 0.2 0.0 0.0 0.1 Congo, Rep. 5,168 0.0 5.6 73.8 20.2 0.0 0.2 0.0 0.0 0.2 Rwanda 1,051 4.4 17.7 21.4 52.8 0.0 3.4 0.0 0.0 0.2 São Tomé and Príncipe 88 0.0 93.2 0.0 6.8 0.0 0.0 0.0 0.0 0.0 Senegal 8,044 5.8 72.2 0.2 3.9 0.0 7.4 9.1 0.0 1.4 Sierra Leone 213 0.0 3.4 0.0 84.4 0.0 9.8 0.0 0.0 2.3 Seychelles 625 0.0 86.4 0.0 0.0 0.0 12.7 1.0 0.0 0.0 Somalia 412 0.0 82.5 0.0 0.0 0.0 16.0 1.5 0.0 0.0 South Africa 220,843 84.3 2.5 0.0 1.2 3.7 2.9 5.3 0.0 0.2 South Sudan 590 0.0 93.2 0.0 0.0 0.0 6.8 0.0 0.0 0.0 Sudan 16,747 0.0 29.9 0.0 68.7 0.0 0.8 0.0 0.0 0.6 Tanzania 10,991 0.0 0.6 73.9 24.6 0.0 0.3 0.0 0.0 0.7 Togo 922 0.0 7.7 71.6 8.7 0.0 11.9 0.0 0.0 0.1 Uganda 5,747 0.0 2.6 0.0 86.6 0.0 2.6 0.0 0.0 8.2 Zambia 19,448 10.9 0.0 0.0 87.9 0.0 0.8 0.0 0.0 0.4 Zimbabwe 8,308 32.5 0.0 0.0 65.7 0.0 0.4 0.0 0.0 1.5 Source: EIA (2025) 14 3.1.3 Capacity factor Capacity utilization is another factor for operational costs because utilities with higher capacity utilization (or capacity factor) can produce more electricity from the same level of installed capacity as compared to those with lower capacity factors. There are two main reasons that affect the capacity factor: generation resource availability and load curve. Renewable energy resources such as hydropower, solar and wind have relatively lower capacity factors as these resources vary across the seasons and time of a day (e.g., solar is not available to generate electricity at night and less hydro is available during the dry season). On the other hand, thermal power plants and nuclear power plants can run all the time except for the time they are stopped for regular maintenance. Load or demand curve varies across hour of a day, weekdays and weekends in a month, months in a season and season in a year. Countries with larger variations of the load curve require more capacity to meet the peak demand than in countries where load variations are lower. For example, in countries where the residential sector is the main consumer of electricity, which is the case in most SSA countries, capacity factors are relatively lower because evening load (due to lighting, cooking) would be much higher than that during the day or night. Figure 5 plots capacity factors of utilities along with the operational costs. From the figure we can notice that utilities which have higher capacity factors have lower operational costs and vice versa. 15 Figure 5. Capacity factor vis-à-vis operational costs 700 90.0% Operating costs (US$/MWh) 600 80.0% 70.0% Capacity Factor (%) 500 60.0% 400 50.0% 300 40.0% 30.0% 200 20.0% 100 10.0% 0 0.0% Angola (PRODEL) Benin (SEBE) Burkina Faso (SONABEL) Cen. Afr. Rep. (ENERCA) Gambia (NAWEC) Malawi (ESCOM) Seychelles (PUC) Botswana (BPC) Mali (EDM) Cote d' Ivoire (CIENERGIES) Mozambique (EDMMOZ) Namibia (NAMPOWER) Niger (NIGELEC) Ruwanda (EUCL) Togo (CEET) Sao Tome & Principe (EMAE) Burundi (REGIDESO) Cabo Verde (ELECTRA) Cameroon (ENEO) Ethiopia (EEP) Guinea (EDG) Liberia (LECLIB) Madagascar (JIRMA) Congo, Dem. Rep. (SNEL) Ghana (VRA) Kenya (KenGen) Senegal (SENELEC) Chad (SNE) Tanzania (TANESCO) Gabon (SEEG) Lesotho (LECLES) Mauritius (CEB) South Africa (ESKOM) Zimbabwe (ZPC & ZESCO) Operational costs Capacity utilization 3.1.4 Capital productivity Lower capital productivity could be another reason for higher operational costs. Figure 6 presents capital productivity and operational costs (including debt service costs). Debt service cost is included because utilities still must pay the debt for the assets they own. Capital productivity is measured by the amount of electricity supplied per unit of asset values. As illustrated in the figure, most countries which have higher capital productivity also have lower operation costs as well as operation plus debt service costs (e.g. South Africa, Mozambique, Mauritius). On the other hand, countries with lower capital productivity also suffer from the higher operational costs as well as operational and debt service costs (e.g., Benin, Botswana, Chad, the Central African Republic, Rwanda, São Tomé & Príncipe). 16 Figure 6. Capital productivity vis-à-vis operational and debt service costs 700 9.00 Operation & debt service costs electrcity in MWh per thousand US$ 600 8.00 Capital productivity (generation of 7.00 500 6.00 400 5.00 4.00 (US$/MWh) 300 asset value) 200 3.00 2.00 100 1.00 0 Cen. Afr. Rep.… 0.00 Sao Tome & Principe… Cote d' Ivoire… Congo, Dem. Rep.… Burkina Faso… Mozambique… Zimbabwe (ZPC &… Benin (SEBE) Burundi (REGIDESO) Cameroon (ENEO) Liberia (LECLIB) Madagascar (JIRMA) Malawi (ESCOM) Chad (SNE) Namibia (NAMPOWER) Niger (NIGELEC) Ruwanda (EUCL) Togo (CEET) Botswana (BPC) Cabo Verde (ELECTRA) Gabon (SEEG) Guinea (EDG) Seychelles (PUC) Mali (EDM) Mauritius (CEB) Senegal (SENELEC) Gambia (NAWEC) Tanzania (TANESCO) South Africa (ESKOM) Operational costs Debt service costs Asset productivity 3.1.5 Labor productivity We also investigate the labor productivity of electric utilities. This is represented by the amount of electricity generated per unit of employment for labor. Figure 7 presents labor productivity of electric utilities in the SSA region. Among the VIUs for which we have data, ESKOM of South Africa has the highest labor productivity as one employee there corresponds to 5.4 GWh of electricity generation, whereas the electric utility in the Central African Republic has the lowest labor productivity. In the case of utilities that are responsible for only distribution (EDUs) or both transmission and distribution (TDUs), Ghana’s electricity distribution utility, ECG, has the highest labor productivity and Nigeria’s AEDC has the lowest. While it would be more relevant to investigate the role of wage expenditure in operational costs, it is not possible to do this due to lack of data. 17 South Africa (ESKOM) 5.4 4.6 (ECG) Ghana Namibia (NAMPOWER) 4.1 Zimbawe (ZESCO) 2.4 DR Congo (SNEL) 2.3 3.1 Angola (ENDE) Mozambique (EDMMOZ) 2.3 Cameroon (ENEO) 2.2 Botswana (BPC) 2.1 2.7 Cote d' Ivoire (CIE) Madagascar (JIRMA) 2.1 Source: World Bank (2025c) Guinea (EDG) 1.9 2.3 Senegal (SENELEC) 1.7 (IKEJA) Nigeria Ruwanda (EUCL) 1.6 Tanzania (TANESCO) 1.4 2.0 Mauritius (CEB) 1.3 Nigeria (EKEDC) Mali (EDM) 1.2 18 Gabon (SEEG) 1.2 1.7 Burkina Faso (SONABEL) 1.2 Uganda Togo (CEET) 1.1 GWh/Employment - VIU Benin (SEBE) 0.9 1.6 Namibia Niger (NIGELEC) 0.8 GWh/Employee - EDU and TDU Chad (SNE) 0.8 Cabo Verde (ELECTRA) 0.5 1.4 Namibia Figure 7. Electricity generation or purchase per unit of employment Liberia (LECLIB) 0.4 (UMEME) (ERONGO) (CENORED) Malawi (ESCOM) 0.4 Seychelles (PUC) 0.3 1.1 Kenya (KPLC) Burundi (REGIDESO) 0.3 1.0 Gambia (NAWEC) 0.2 Sao Tome & Principe… 0.2 Nigeria (AEDC) import. ETU and EDU purchase electricity from EGUs and IPPs. EDU can have their own generation as well. Cen. Afr. Rep. (ENERCA) 0.2 Note: VIUs can generate electricity by themselves, purchase from independent power producers (IPPs) and 3.1.6 Debt burden Does higher debt cause higher operation and debt service costs? We also investigate this question by plotting the debt-to-asset ratio vis-à-vis operation and debt service costs (Figure 8). Most utilities exhibit higher operation and debt service costs if they have a higher debt-to-asset ratio. There are, however, some exceptions. For example, VIUs in Chad, Guinea, Lesotho, Liberia, and São Tomé and Príncipe have higher operation and debt service costs despite their lower debt-to-service ratios. The reverse is the case in Côte d’Ivoire and South Africa. As far as other utilities are concerned, generation-only utilities have higher debt-to-asset ratios and lower operational and debt service costs because of the high debt they borrowed during the construction of power plants, but their operational costs are lower because they use renewable energy (e.g., hydro, geothermal) which does not have fuel costs, for power generation. Figure 8. Operation and debt service cost vis-à-vis debt to asset ratio 700 1.200 Operational cost Debt service cost Debt to assets ratio Operation & debt service costs (US$/MWh) 600 1.000 500 0.800 400 0.600 300 0.400 200 100 0.200 0 0.000 Burkina Faso… Congo, Dem. Rep.… Cote d' Ivoire… Sao Tome & Principe… Benin (SEBE) Liberia (LECLIB) Madagascar (JIRMA) Namibia (NAMPOWER) Seychelles (PUC) Zimbawe (ZESCO) Botswana (BPC) Cen. Afr. Rep. (ENERCA) Chad (SNE) Cameroon (ENEO) Burundi (REGIDESO) Gabon (SEEG) Guinea (EDG) Niger (NIGELEC) Lesotho (LECLES) Mauritius (CEB) Ruwanda (EUCL) Malawi (ESCOM) Mali (EDM) Cabo Verde (ELECTRA) Gambia (NAWEC) Mozambique (EDMMOZ) Senegal (SENELEC) South Africa (ESKOM) Togo (CEET) Tanzania (TANESCO) VIU 19 300 0.500 Operation & debt service costs (US$/MWh) Operation cost Debt service cost Debt to assets ratio 0.450 250 0.400 200 0.350 0.300 150 0.250 0.200 100 0.150 50 0.100 0.050 0 0.000 Ethiopia (EEU) Nigeria (YEDC) Cote d' Ivoire (CIE) Namibia (ERONGO) Nigeria (AEDC) Ethiopia (EEP) Ghana (VRA) Angola (ENDE) Ghana (ECG) Nigeria (KEDCO) Nigeria (PHED) Angola (RNT) Uganda (UETCL) Kenya (KenGen) Zimbabwe (ZPC) Namibia (CENORED) Uganda (UMEME) Kenya (KPLC) Nigeria (EKEDC) Nigeria (IKEJA) Angola (PRODEL) EDU TU TDU EGU 3.2 Revenue factors 3.2.1 Electricity tariffs Most SSA and many other developing countries' governments set electricity tariffs. Electricity tariffs are designed considering several factors, such as cost recovery, vertical equity (affordability) and horizontal equity (price differentiation) and incentives for adoption of emerging technologies (e.g., distributed generation and storage, electric vehicles, and demand-side participation) (Mburamatare et al. 2022; Foster and Witte, 2020). Are electricity tariffs so low in SSA countries that they cause the utilities to be unable to cover their production cost (in the case of VIUs) or purchasing cost (in the case of EDUs)? Often, electricity tariffs are blamed for the poor financial performance of utilities and tariff hikes are recommended (see for example, Burgess et al. 2020; Kojima and Trimble, 2016). We investigate this question by comparing the most recent electricity prices in SSA countries with those of countries in other regions. Figure 9 plots electricity prices against per capita GDP (both expressed in 2015 constant US$) for countries in different regions for which data are available. It is interesting to note that electricity prices in Sub-Saharan African countries are at the higher end. Electricity prices in most SSA countries are higher than those in most 20 countries in other regions except OECD countries. They are also higher than those in some OECD countries: Canada, Hungary, Iceland and Türkiye. The average per capita income (GDP) of the SSA countries included in Figure 9 is US$2,058, whereas Canada’s per capita GDP is US$44,469, almost 22 times higher. The average electricity prices of ECA and EAP countries are, respectively, US$83 and US$100 per MWh, whereas the corresponding value for SSA is US$151/MWh (see lower panel of Figure 9). On the other hand, ECA and EAP countries have about 3.5 to 4 times higher income per capita as compared to SSA countries. Based on Figure 9, it would be difficult to argue that electricity tariffs in SSA countries are lower. If affordability (measured by per capita GDP) is accounted for, their electricity tariffs are the highest in the world. It is possible that they are lower than their production costs, but this does not mean that tariffs should be increased further to equalize production costs. Instead, policies must be focused on reducing production costs as discussed earlier. 5 Although some restructuring in tariff structure is possible by setting a higher tariff for the rich (those who consume a higher quantity of electricity) and lower tariff for the poor, increasing the electricity tariff for all consumers may not be justified from an equity perspective. Moreover, hiking electricity tariffs is politically sensitive and could face severe resistance. In some cases, demonstrations were triggered by an electricity price increase. 6 It does not, however, mean that governments should continue the existing direct subsidies to electric utilities. They should first implement all possible ways to reduce the supply cost before considering tariff hikes. Moreover, if the electricity tariff is lower than operational costs in countries where operational costs are already relatively low, careful tariff reform (e.g., rate hike, removal of subsidies) protecting the poor (e.g., through a lifeline tariff) can be considered. 5 We have also plotted using nominal prices, the figure does not change much. 6 See, for example in Puerto Rico in July 2023 (https://apnews.com/article/puerto-rico-protest-power-bills- 87e7c6f0c6a83e06aea67c133de93f52AP). 21 Per capita GDP ('000 US$) Per capita GDP ('000 US$) 0 10 20 30 40 50 60 70 80 90 0.0 10.0 20.0 30.0 40.0 50.0 Australia Canada Chile Croatia Czechia Estonia OECD Finland France Germany Greece Hungary Iceland Israel EAP Japan OECD Korea Malta Netherlands New Zealand Norway Portugal Türkiye Figure 9. Electricity prices vs. per capita GDP ECA Singapore Spain Sources: IEA (2025) and World Bank (2025a) Switzerland United Kingdom United States China Indonesia Malaysia EAP LAC Mongolia 22 GDP Per Capita ('000 US$) Thailand Azerbaijan Regional average values Belarus Georgia Kazakhstan Kosovo ECA Montenegro Individual countries MENA Russia Serbia Average prices GDP Per Capita ('000 US$) Uzbekistan Brazil Colombia Costa Rica LAC Nicaragua Paraguay SA Algeria Egypt Jordan MENA Tunisia India Electricity Price Nepal SA Pakistan Benin SSA Chad Cote d'Ivoire Ghana SSA Kenya 0 50 Senegal 100 150 200 250 South Africa Electricity price (US$/MWh) Uganda 0 50 100 150 200 250 300 350 400 450 Electricity price (US$/MWh) 3.2.2 Transmission and distribution (T& D) losses T&D losses are one of the primary factors contributing to the poor financial performance of utilities not only in SSA but also other regions (Galeazzi et al. 2024; Burgess et al. 2020; Trimble et al. 2016). Figure 10 presents electricity system losses 7 of VIUs and electricity distribution losses of EDUs. 8 As can be seen from the left panel of Figure 10, electricity system losses of VIUs vary from 6.2% (Mauritius) to 38% (Cameroon). Although it varies across the countries, electricity T&D losses of efficient electric utilities stand below 10%. The U.S. electric utilities report about 5% T&D (EIA, 2024). Being a small network system, Singapore’s electricity system is considered one of most efficient electric utility systems in the world with T&D losses about 2%. Even in the SSA region, T&D losses in Mauritius and the Seychelles are less than 7% due to efficient supply systems and relatively shorter T&D networks (Figure 10). In the case of EDUs, data on distribution losses are mostly missing. In a few countries, data on distribution losses are available but they are almost 10 years old. We later estimate improvements in the financial performance of utilities (revenue- to-operational cost ratios) if T&D losses are reduced and brought to a level reasonable for SSA utilities. Unlike the case of tariff hikes, investments in reducing T&D losses do not face any political backlash. It is one of the most preferred options to improve the financial performance of electric utilities (Burgess et al. 2020; Kozima et al. 2016; Trimble et al. 2016). 3.2.3 Electricity bill collection rate Revenue loss due to lower bill collection rate is another factor that reduces the revenue base of electric utilities in SSA and is thus responsible for their poor financial performance. Figure 11 (left panel) shows the collection rates for VIUs. About one-third of VIUs have collection rates below 90%. Four VIUs collect less than 60% of their electric bills. In the case of other types of utilities (EGUs, ETUs, TDUs and EDUs), the situation is even worse. Of the 26 utilities for which we have data, more than half (14) lose more than 10% of 7 Electricity system losses account for transmission losses, distribution losses and own-use consumption in power plants. 8 The losses are relatively small for EGUs and ETUs and not discussed here. 23 revenue due to collection leakage (Figure 11, right panel). Eight of these utilities lose more than 40% of their revenue due to collection losses. We also calculate later the degree of improvement of financial performance of electric utilities (all types) in the SSA region if collection loss is eliminated. While elimination of collection loss requires big efforts by utilities and may take many years, our objective here is to understand the maximum increase in revenue if the collection loss is eliminated. Figure 10. T&D loss of electric utilities in the SSA region (% of production or supplied to the T&D networks) Cameroon (ENEO) 38.0 Chad (SNE) 37.0 Liberia (LECLIB) 32.0 Congo, Dem. Rep. (SNEL) 31.2 Cen. Afr. Rep. (ENERCA) 28.0 Malawi (ESCOM) 27.4 Mozambique (EDM-MOZ) 26.0 Sao Tome & Principe (EMAE) 24.3 Gabon (SEEG) 24.0 Cabo Verde (ELECTRA) 24.0 Benin (SEBE) 24.0 Niger (NIGELEC) 23.5 Mali (EDM) 23.4 Mauritania (SOMELEC) 22.9 Burundi (REGIDESO) 19.2 Guinea (EDG) 19.0 Senegal (SENELEC) 17.7 Gambia (NAWEC) 17.6 Ruwanda (EUCL) 16.9 Togo (CEET) 16.0 Zimbabwe (ZESCO) 16.0 Burkina Faso (SONABEL) 15.5 Madagascar (JIRMA) 15.4 Botswana (BPC) 15.4 Cote d' Ivoire (CIENERGIES) 15.0 Tanzania (TANESCO) 14.5 Lesotho (LECLES) 13.2 South Africa (ESKOM) 11.8 Namibia (NAMPOWER) 11.8 Seychelles (PUC) 7.0 Mauritius (CEB) 6.2 Vertically integrated utility (VIU) Source: World Bank (2025c) 24 Figure 11. Electricity bills collection rates (%) Niger (NIGELEC) 100 Zimbabwe (ZPC) 97 Namibia (NAMPOWER) 100 Uganda (UEGCL) 47 Mozambique (EDMMOZ) 100 Sudan (STPGC) 86 Mauritius (CEB) 100 Sudan (SHREG) 98 EGU South Africa (ESKOM) 99 Kenya (KenGen) 99 Ghana (VRA) 100 Cabo Verde (ELECTRA) 99 Ethiopia (EEP) 80 Burkina Faso (SONABEL) 98 Angola (PRODEL) 59 Tanzania (TANESCO) 98 Zimbabway (ZETDC) 77 Seychelles (PUC) TDU 97 Kenya (KPLC) 98 Lesotho (LECLES) 97 Cote d' Ivoire (CIE) 87 Mali (EDM) 96 Uganda (UETCL) 92 Sudan (SETC) 77 Cameroon (ENEO) 96 Nigeria (TCN) 52 ETU Gambia (NAWEC) 96 Kenya (KETRACO) Zimbawe (ZESCO) 96 Ghana (GRIDCO) 43 Togo (CEET) 96 Angola (RNT) 68 Burundi (REGIDESO) 92 Uganda (UMEME) Madagascar (JIRMA) 91 Namibia (NORED) 100 Botswana (BPC) Namibia (ERONGO) 99 91 Nigeria (KEDCO) 99 Senegal (SENELEC) 91 Namibia (CENORED) 97 Gabon (SEEG) 91 Sudan (SEDC) 97 Malawi (ESCOM) 90 Ethiopia (EEU) 95 Mauritania (SOMELEC) 89 Nigeria (IKEJA) 92 Benin (SEBE) 87 Nigeria (AEDC) 92 EDU Ruwanda (EUCL) 85 Nigeria (EKEDC) 92 Sierra Leone (EDSA) 81 Liberia (LECLIB) 82 Nigeria (JOS) 80 Sao Tome & Principe (EMAE) 78 Angola (ENDE) 70 Chad (SNE) 64 Nigeria (IBEDC) 60 Cen Afr Rep (ENERCA) 56 Nigeria (BEDC) 56 Cote d' Ivoire (CIENERGIES) 55 Nigeria (YEDC) 55 VIU DR Dem (SNEL) 54 Nigeria (PHED) 44 Guinea (EDG) Ghana (ECG) 40 36 Source: World Bank (2025c) 25 Most existing studies are found to focus on increasing revenue rather than reducing production costs to improve the financial performances of utilities. Reviewing 82 existing studies on the impacts of electricity tariff reform in Africa, Klug et al. (2022) find that prepaid meters help reduce leakages in electricity bill collection, thereby increasing the revenues of electric utilities. Jack and Smith (2020) experiment on this hypothesis using data collected from 4,000 residential customers in Cape Town, South Africa and find that switching to prepaid metering increases utilities’ overall revenues although electrcity consumption drops by 14%. Burgess et al. (2020) also stress increasing bill collection through smart metering, increasing customer awareness and social trust mechansim to increase utilities’ revenue base. There are some other factors that might have caused higher costs and lower revenue and are not reflected in the indicators discussed above. These include political economy and institutional and governance issues (Yeo, 2024; Twesigye, 2024; Todd and Bazilian, 2018; Auriol and Blanc, 2009). Many electric utilities face overspending on operation and maintenance costs due to poor governance (Trimble et al. 2016). Project delays and cancellations are common phenomena which lead to cost overruns and higher production costs due to increased interest payments during construction and debt payments during operation (Todd and Bazilian, 2018). Weaker institutional capacity and lack of proper regulatory arrangements for management of project contracts also contribute to higher costs of electricity supply (Twesigye, 2024). Political interests could lead to a selection of projects which are not necessarily the least cost options, thereby increasing the cost of electricity supply (Yeo, 2024). If these issues are addressed, it would help to reduce production costs as well as to increase revenue. However, no empirical study is available estimating reduction in production costs and increased in revenues of electric utilities through these factors for the SSA region. 26 4. Improving the financial performance of utilities in the SSA region 4.1 Revenue channels 4.1.1 Improving financial performance through T&D loss reduction We estimate here how much the average revenue per unit of electricity sold could be increased if electricity system losses of SSA VIUs improved to the level of South Africa’s current system losses, which is 11.8%. This level of system losses indicates an efficient power system in terms of system losses considering the area of South Africa and its generation mix. Figure 12 (upper panel) presents the percentage of operational costs that would be covered by revenues when system losses are decreased to the level of South Africa’s current system losses. The reduction of system losses would increase their revenues by 1% to 36% depending upon the level of their current system losses. However, only three VIUs out of 15 VIUs, which do not cover their operational costs by their revenue at present, are able to do so if the increased revenue caused by technical loss reduction is accounted for. 9 A few existing studies have also attempted to estimate the increased revenue through electricity loss reduction. For example, Gautier et al. (2023) estimate that East African countries could generate US$60 million per year by reducing their T&D losses by 8% in 10 years. 4.1.2 Improvement of financial performance through elimination of collection losses We then calculate the increased revenue if electricity bills are collected 100%. Figure 12 (upper panel) shows that revenues of VIUs would have increased from 1% to 181% if there were no collection losses. Four VIUs, which have revenue lower than operational cost, would realize more revenue than their operational cost if there were no losses in bill collection. Two more VIUs could have higher revenues than operational costs if system losses are reduced and all electricity bills are collected. Altogether nine VIUs out of 15 would 9 Due to lack of data, we are unable to calculate the increased revenues with distribution loss reduction for EDUs. 27 50 100 150 200 250 300 350 400 0 100 200 300 400 500 600 700 0 Angola (ENDE) Benin (SEBE) Ethiopia (EEU) Botswana (BPC) Ghana (ECG) Burkina Faso… Burundi (REGIDESO) Namibia (CENORED) Cabo Verde… Namibia (ERONGO) Cen Afr Rep… Namibia (NORED) Chad (SNE) loss Nigeria (AEDC) DR Congo (SNEL) eliminating collection losses. Gabon (SEEG) EDU Nigeria (EKEDC) Gambia (NAWEC) Base revenue VIU Nigeria (IKEJA) Guinea (EDG) Nigeria (KEDCO) Lesotho (LECLES) Liberia (LECLIB) Base revenue Nigeria (PHED) Madagascar (JIRMA) Operational cost eliminating bill collection loss (US$/MWh) Nigeria (YEDC) Malawi (ESCOM) 28 Sierra Leone (EDSA) Mali (EDM) Uganda (UMEME) Mauritius (CEB) Mozambique… Angola (RNT) Namibia… Source: Author’s calculation based on data from World Bank (2025c) Ghana (GRIDCO) ETU Niger (NIGELEC) Uganda (UETCL) Ruwanda (EUCL) Cote d' Ivoire (CIE) Senegal (SENELEC) Seychelles (PUC) EDT Incremental revenue through T&D loss reduction Kenya (KPLC) South Africa (ESKOM) Angola (PRODEL) Sao Tome &… Incremental revenue through elimination of elctricity bill collection Ethiopia (EEP) Tanzania (TANESCO) Incremental revenue through bill collection loss elimination EGU Ghana (VRA) Togo (CEET) Zimbawe (ZESCO) Kenya (KenGen) Cameroon (ENEO) 0 0 50 100 150 200 250 300 350 400 100 200 300 400 500 600 700 Figure 12. Improvement of financial performance of utilities by reducing T&D loss and achieve more revenues than operational costs through the reduction of system losses and We also calculate the increase in revenues through the elimination of collection losses in other types of utilities (EGUs, ETUs, TDUs and EDUs) for which necessary data are available. Note that we could not calculate an increase in revenue by reducing T&D losses due to lack of data for these utilities. Figure 12 (lower panel) shows that 10 electric utilities have lower revenue than operational costs with existing collection losses. If the collection losses are eliminated four out of these 10 utilities would realize higher revenue than their operational costs. Despite the reduction of T&D losses and elimination of collection losses, 8 VIUs out of 30 still have lower revenue than their operational costs. Similarly, six other utilities (EGUs, ETUs, TDUs, EDUs) out of 26 utilities are still in losses (i.e., revenue is lower than operational cost) even if their collection losses are eliminated. For these utilities other avenues, such as improving productivity, should be explored. However, we could not calculate reduction in operational costs due to an increase in labor and capital productivity due to the lack of data. 4.2 Cost channels Unlike the case of revenue channels, reducing operational costs through the cost channel is difficult as it requires substantial long-term investments. A country’s electricity generation mix is normally determined through electricity capacity expansion planning based on least cost modeling. Such modeling determines investments in a series of new power plants over a longer time horizon considering the operation costs of both existing and new power plants and capacity expansion costs of new power plants. One possible option could be, especially in those countries where expensive imported diesel fuel-based power plants predominate electricity generation, should consider solar power plants when adding new generation capacities, and utilizing the existing diesel-fired power plants for back-up generation. This is because solar plants are getting cheaper due to rapid decline in costs and increasing efficiency. It could reduce fuel costs and the overall operational cost. Solar plants have capacity costs as low as US$600/kW (IRENA, 20024; Timilsina, 2021) and the levelized costs are one of the lowest US$22/MWh in regions with moderate discount rate 29 (7%) and high solar irradiation (most of the SSA region). For the same discount rate, the levelized cost of the cheapest fossil-fuel based power generation (natural gas combined cycle) is estimated to be US$39/MWh at a lower gas price of US$3.3/GJ (Timilsina, 2021). Diesel-fired internal combustion engines have minimum capital costs of US$500/kW and corresponding levelized costs would be US$682/MWh with 28% thermal efficiency and US$2/liter diesel price on average delivered in the SSA region in 2023. 10,11 It implies that electricity supply from solar PV plants would be much cheaper than that from diesel-fired power plants even though the former has much lower capacity factor (30%) than the latter (85%). Moreover, solar power plants can be installed within a year, whereas the installation of other conventional power plants (e.g., hydro, thermal) takes several years. It not only saves interest payments during the construction but also makes electricity available sooner. Increasing capital and labor productivity would also lower the costs of electricity production. Capital cost could be a major concern especially in those countries where oil- based power generation predominates (e.g., Cabo Verde, Chad, The Gambia, Mali, São Tomé and Príncipe, Senegal). Since the capital costs of oil-fired power plants are relatively lower and thus the lower asset values, having lower capital productivity (i.e., lower electricity generation per unit of asset value) indicates that there might be other physical assets (e.g., vehicles, buildings) being used more inefficiently. Thus, reducing or selling off other unnecessary physical assets could help reduce operational costs. 5. Conclusions and Policy Implications Electric utilities are the main providers of electricity services in Sub-Saharan Africa. Most of the electric utilities in the region are state-owned and have a vertically integrated structure with generation, transmission and distribution responsibility. A few countries have restructured the electricity sector, thereby creating generation, transmission and 10 https://www.statista.com/statistics/1297071/average-retail-prices-for-diesel-in-africa-by-country/ 11 Author’s calculation. 30 distribution or their combinations as separate utilities. A few distribution utilities have been privatized. This study first analyzed the financial and technical performance of electric utilities in the Sub-Saharan Africa region using data for 67 electric utilities in 47 countries available from various sources including the World Bank, International Energy Agency, United States Energy Information Administration and utilities annual reports. The study found that 16 vertically integrated utilities of 30 cannot cover their operational costs by revenues, thereby running at a loss. If the cost of debt is also included, another four VIUs are operating at a loss. In some cases, operational costs are more than twice as high as their revenues. Electricity distribution utilities are not doing any better. Only five of the 19 electricity distribution companies considered in the study were found to have revenues higher than their operational and debt costs. Generation only and transmission only utilities were found to have higher revenues than their operational costs because electricity transmission systems have relatively lower operational costs. Electricity generation utilities with predominantly hydropower assets have relatively low operational costs in the absence of fuel costs for power generation. The study foung that both cost-side and revenue-side factors are responsible for the poor financial performance of electric utilities in the Sub-Saharan Africa region. The factors causing higher operational costs were found to be higher fuel costs, lower capacity factors, lower capital and labor productivity. Utilities operating with higher losses were found to predominantly use oil for electricity generation, particularly diesel. Oil-based power plants have much higher electricity production costs due to low thermal efficiency and high fuel costs. Besides the utilities with predominantly hydro-based power generation, utilities that predominantly use natural gas were also found to have low operational costs. This is because domestic natural gas is cheaper in the region due to lack of export infrastructure. Factors that contribute to lower revenues are high T&D losses and lower rates of electricity bill collection. Most utilities have relatively high T&D losses, more than 30% in some cases. Higher T&D losses mean lower sales of electricity generated or purchased. Most utilities were found to be unable to fully collect their electricity bills, and some have 31 bill collection rates below 50%. Electricity tariffs are another critical factor for the level of revenue a utility generates. Comparing the average residential electricity tariffs of selected African countries for which data is available with those of countries in other regions, we found that electricity tariffs in Sub-Saharan African countries are higher than the tariffs in many non-OECD countries and even some OECD countries. Considering that the per capita GDP of Sub-Saharan African countries is several times lower than that of OECD, MENA, EAP and ECA countries, Sub-Saharan African countries are already paying relatively higher prices for electricity despite their low per capita income. The study also estimated how much of the financial gaps can be lowered by improving technical performance, particularly by reducing T&D losses and by increasing bill collection rates. If the electricity system losses of vertically integrated utilities were reduced to the current level of South Africa’s system losses, three vertically integrated utilities out of 15, which currently do not cover their operational costs by their revenue, could do so. If the electricity bills were fully collected, nine vertically integrated utilities out of 15 would achieve more revenues than operational costs through the reduction of system losses and elimination of collection losses. Switching to utility scale solar photovoltaics could be another strategy to reduce operational costs for those countries where oil-based generation is currently predominant. Significant drops in capital costs, short installation period, zero fuel cost and high solar irradiation (solar energy resources) could make solar PV the best alternative to expensive oil-based power generation in these countries. Moreover, existing oil-based power plants could be used for back-up purposes when solar energy is not available. Country-specific new analyses might be helpful to explore this option further. The lower capital productivity of electric utilities in the SSA region also implies overspending on or misuse of vehicles and buildings that are owned by utilities. Reducing such misuse would increase capital productivity. There are some other reasons behind the higher operation and debt service costs. These include overspending on construction and purchase of equipment, cost overruns due to project delays, cancellation and inefficient management of contracts and other institutional inefficiencies. Further analysis is needed 32 to understand how much these factors have contributed to the poor financial performance of electric utilities in the Sub-Saharan Africa region. Acknowledgments This paper is produced through research activities under the Africa Energy Program of the Institute for Economic Development, Development Economics Vice Presidency, World Bank, Washington, DC. The authors would like to acknowledge Carolyn Fischer, Mike Toman, Debabrata Chattopadhyay, Jayas Poudel and Ashish Shrestha for their constructive comments. The views and interpretations are of the authors and should not be attributed to the World Bank Group and the organizations they are affiliated with. References Auriol, Emmanuelle and Aymeric Blanc (2009). Capture and corruption in public utilities: The cases of water and electricity in Sub-Saharan Africa. Utilities Policy, Vol. 17, pp. 203– 216. Balabanyan, A., Y. Semikolenova, A. Singh, and M. A. Lee. 2021. Utility Performance and Behavior in Africa Today (UPBEAT): SUMMARY REPORT. ESMAP Papers. World Bank, Washington, DC. Burgess, Robin, Michael Greenstone, Nicholas Ryan, and Anant Sudarshan (2020), The Consequences of Treating Electricity as a Right. Journal of Economic Perspectives, Volume 34, Number 1, Pages 145–169. Energy Information Administration (EIA) (2025). Online Energy Database. United States Department of Energy. https://www.eia.gov/international/data/world/electricity/ Foster, Vivien and Samantha Witte (2020). Falling Short: A Global Survey of Electricity Tariff Design. World Bank Policy Research Working Paper 9174, World Bank, Washington, DC. Fritsch, R (2011). Comparative financial analysis of electricity utilities in West Africa, Energy Policy, Vol. 39, pp. 6055–6064. Galeazzi, Clara, Jevgenijs Steinbuks and Laura Diaz Anadon (2024). Assessing the impact of renewable energy policies on decarbonization in developing countries. Renewable and Sustainable Energy Reviews, Volume 199, 114444. 33 Gautier, Axel, René Nsabimana and Barnabé Walheer (2023). Quality performance gaps and minimal electricity losses in East Africa. Utilities Policy, Vol. 82; 101558. Hafner, Manfred, Simone Tagliapietra and Lucia de Strasser (2018). Energy in Africa: Challenges and Opportunities. Spring Book Open. https://doi.org/10.1007/978-3- 319-92219-5 International Energy Agency (IEA) (2022). Africa Energy Outlook 2022. IEA, Paris. International Energy Agency (IEA) (2025). Online Energy Database. https://www.iea.org/data-and-statistics. Jack, Kelsey and Grant Smith (2020). Charging Ahead: Prepaid Metering, Electricity Use, and Utility Revenue. American Economic Journal: Applied Economics, Vol. 12(2), pp. 134–168. Klug, Thomas W., Abebe D. Beyene, Tensay H. Meles, Michael A. Toman, Sied Hassen, Michael Hou, Benjamin Klooss, Alemu Mekonnen, Marc Jeuland (2022), A review of impacts of electricity tariff reform in Africa. Energy Policy, Vol. 170; 113226. Kojima, Masami and Chris Trimble (2016). Making Power Affordable for Africa and Viable for Its Utilities. World Bank, Washington, DC. Mburamatare, Daniel, William K. Gboney, Jean De Dieu Hakizimana (2022). Electricity Tariff Design “Theoretical Concepts versus Practices”: Review of Tariff Design Approaches in East Africa-Case Studies of Rwanda, Kenya, Uganda and Tanzania. International Journal of Energy Economics and Policy, Volume 12(5), pp. 260-273. Moss, Todd and Morgan Bazilian (2018). Signaling, governance, and goals: Reorienting the United States Power Africa initiative. Energy Research & Social Science, Vol. 39 pp. 74–77. Rana, Anshul, Mumba Ngulube and Vivien Foster (2022). GERI 2022: Global Electricity Regulatory Index. World Bank, Washington, DC. Twesigye, Peter (2024). Understanding Electricity Utilities in Sub-Saharan Africa: The Role of Civil Society in Improving Performance, Governance, and Accountability Relationships. London: Oxfam. Trimble, Christopher Philip, Masami Kojima and Arroyo Ines Perez and Farah Mohammadzadeh (2016). Financial Viability of Electricity Sectors in Sub-Saharan 34 Africa: Quasi-Fiscal Deficits and Hidden Costs. World Bank Policy Research Working Paper No. 7788, World Bank, Washington, DC. United Nations. Economic Commission for Africa (UNECA) (2023). African Countries pay more to borrow. ECA Policy Brief. ECA/23/031, UNECA, Addis Ababa. https://hdl.handle.net/10855/49851 World Bank (2025a). World Development Indicator Online Database. https://databank.worldbank.org/source/world-development-indicators. World Bank (2025b). World Bank Business Ready (B-READY) Online database. https://www.worldbank.org/en/businessready World Bank (2025c). World Bank’s Utility Performance and Behavior Today Database. https://utilityperformance.energydata.info/ Yeo, Wondia Mireille (2024). A political economy analysis of the electricity sector in Cote d'Ivoire. Energy Research & Social Science 118 (2024) 103833. 35 Appendix Table A1: Electricity Utilities in Sub-Saharan Africa (a) Utilities with vertically integrated monopoly structure (the same utility provides electricity generation, transmission and distribution functions) S.N. Country (Utility) Ownership 1 Benin (SEBE) State owned (Public) 2 Botswana (BPC) State owned (Public) 3 Burkina Faso (SONABEL) State owned (Public) 4 Burundi (REGIDESO) State owned (Public) 5 Cabo Verde (ELECTRA) State owned (Public) 6 Cameroon (ENEO) Private 7 Central African Republic (ENERCA) State owned (Public) 8 Chad (SNE) State owned (Public) 9 Congo, Dem. Rep. (SNEL) State owned (Public) 10 Côte d'Ivoire (CIENERGIES) State owned (Public) 11 Gabon (SEEG) State owned (Public) 12 Gambia, The (NAWEC) State owned (Public) 13 Guinea (EDG) State owned (Public) 14 Lesotho (LECLES) State owned (Public) 15 Liberia (LECLIB) State owned (Public) 16 Madagascar (JIRMA) State owned (Public) 17 Malawi (ESCOM) State owned (Public) 18 Mali (EDM) State owned (Public) 19 Mauritania (SOMELEC) State owned (Public) 20 Mauritius (CEB) State owned (Public) 21 Mozambique (EDM-MOZ) State owned (Public) 22 Namibia (NAMPOWER) State owned (Public) 23 Niger (NIGELEC) State owned (Public) 24 Rwanda (EUCL) State owned (Public) 25 Senegal (SENELEC) State owned (Public) 26 Seychelles (PUC) State owned (Public) 27 South Africa (ESKOM) State owned (Public) 28 São Tomé and Príncipe (EMAE) State owned (Public) 29 Tanzania (TANESCO) State owned (Public) 30 Zimbabwe (ZESCO) State owned (Public) 36 (b) Utilities with electricity generation S.N. Country (Utility) Ownership 1 Angola (PRODEL) State owned (Public) 2 Ethiopia (EEP) State owned (Public) 3 Ghana (VRA) State owned (Public) 4 Kenya (KENGEN) State owned (Public) 5 Sudan (SHREG) State owned (Public) 6 Sudan (STPGC) State owned (Public) 7 Uganda (UEGCL) State owned (Public) 8 Zimbabwe (ZPC) State owned (Public) (c) Utilities with electricity transmission responsibility only S.N. Country (Utility) Ownership 1 Angola (RNT) State owned (Public) 2 Ghana (GRIDCO) State owned (Public) 3 Kenya (KETRACO) State owned (Public) 5 Nigeria (TCN) State owned (Public) 6 Sudan (SETC) State owned (Public) 7 Uganda (UETCL) State owned (Public) (d) Utilities with electricity transmission and distribution (T&D) responsibility S.N. Country (Utility) Ownership 1 Cote d' Ivoire (CIE) Private 2 Kenya (KPLC) State owned (Public) 3 Zimbabwe (ZETDC) State owned (Public) (e) Utilities with electricity distribution responsibility only S.N. Country (Utility) Ownership 1 Angola (ENDE) State owned (Public) 2 Ethiopia (EEU) State owned (Public) 3 Ghana (ECG) State owned (Public) 4 Namibia (CENORED) State owned (Public) 5 Namibia (ERONGO) State owned (Public) 6 Namibia (NORED) State owned (Public) 7 Nigeria (AEDC) Private 8 Nigeria (BEDC) Private 9 Nigeria (EKEDC) Private 10 Nigeria (IBEDC) Private 37 11 Nigeria (IKEJA) Private 12 Nigeria (JOS) Private 13 Nigeria (KEDCO) Private 14 Nigeria (PHED) Private 15 Nigeria (YEDC) Private 16 Namibia (CENORED) State owned (Public) 17 Sierra Leone (EDSA) State owned (Public) 18 Sudan (SEDC) State owned (Public) 19 Uganda (UMEME) Private 38 Figure A1. Size indicators of electric utilities (a) Total asset values (Million US$) 46,563 South Africa… Kenya (KPLC) 3,080 DR Congo (SNEL) 10,037 TDU Cote d' Ivoire (CIE) 2,953 Tanzania… 9,286 Zimbabway (ZETDC) 1,655 Cote d' Ivoire… 6,530 Nigeria (TCN) 2,371 Zimbawe (ZESCO) 4,366 Mozambique… 4,251 Kenya (KETRACO) 2,028 Senegal (SENELEC) 3,013 Angola (RNT) 1,486 ETU Namibia… 2,625 Uganda (UETCL) 1,386 Madagascar… 2,478 Ghana (GRIDCO) 1,164 Botswana (BPC) 2,478 Sudan (SETC) VIU (Vertically integrated Utilities) 117 Cameroon (ENEO) 1,529 Ethiopia (EEU) 1,753 Burkina Faso… 1,184 Angola (ENDE) 1,477 Mauritania… 1,021 Mauritius (CEB) Ghana (ECG) 1,472 1,010 Mali (EDM) 979 Nigeria (IBEDC) 1,381 Ruwanda (EUCL) 924 Nigeria (BEDC) 912 Benin (SEBE) 842 Nigeria (PHED) 702 Malawi (ESCOM) 752 Uganda (UMEME) 692 Guinea (EDG) 642 Nigeria (AEDC) 618 Liberia (LECLIB) 609 Nigeria (IKEJA) 550 EDU Niger (NIGELEC) 571 Nigeria (EKEDC) 474 Togo (CEET) 494 Seychelles (PUC) 396 Nigeria (YEDC) 423 Chad (SNE) 352 Sudan (SEDC) 300 Lesotho (LECLES) 334 Nigeria (KEDCO) 247 Gabon (SEEG) 284 Namibia (NORED) 185 Cabo Verde… 203 Nigeria (JOS) 166 Burundi… 193 Namibia (ERONGO) 142 Gambia (NAWEC) 193 Sierra Leone (EDSA) 56 Central African… 110 Namibia (CENORED) 52 Sao Tome &… 97 Generation utilities (EGU) EGT Ethiopia (EEP) 12,238 Kenya (KenGen) 3,676 Ghana (VRA) 3,364 Zimbabwe (ZPC) 2,524 Angola (PRODEL) 2,221 Uganda (UEGCL) 1,900 Sudan (SHREG) 163 Sudan (STPGC) 68 39 Source: World Bank (2025c) (b) Employment (number of workers) 39,601 South Africa (ESKOM) DR Congo (SNEL) 9,652 Angola (PRODEL) 2693 Zimbawe (ZESCO) 6,765 Zimbabwe (ZPC) 2647 EGU Tanzania (TANESCO) 6,725 Kenya (KenGen) 2593 Mozambique (EDM-MOZ) 4,065 Ghana (VRA) 2086 Cameroon (ENEO) 3,600 Uganda (UEGCL) 200 Senegal (SENELEC) 3,394 10,018 Kenya (KPLC) Burkina Faso (SONABEL) 2,700 TDU Mali (EDM) 2,337 Cote d' Ivoire (CIE) 4,994 Mauritius (CEB) 2,230 Angola (RNT) 1,125 ETU Gabon (SEEG) 1,982 Uganda (UETCL) 476 Madagascar (JIRMA) 1,916 Ghana (ECG) 6,422 Botswana (BPC) 1,916 Sudan (SEDC) 6,094 Benin (SEBE) 1,849 Gambia (NAWEC) 1,815 Angola (ENDE) 4,890 VIU (Vertically integrated utilities) Niger (NIGELEC) 1,780 Nigeria (AEDC) 3,428 Togo (CEET) 1,524 Nigeria (IBEDC) 2,675 Guinea (EDG) 1,487 Nigeria (JOS) 2,622 Burundi (REGIDESO) 1,328 EDU Uganda (UMEME) 2,510 Seychelles (PUC) 1,226 Nigeria (PHED) 2,226 Namibia (NAMPOWER) 1,077 Liberia (LECLIB) 925 Nigeria (IKEJA) 2,016 Cabo Verde (ELECTRA) 826 Nigeria (EKEDC) 2,002 Central African Rep. (ENERCA) 750 Nigeria (YEDC) 930 Chad (SNE) 690 Namibia (ERONGO) 327 Cote d' Ivoire (CIENERGIES) 580 Namibia (CENORED) 165 Sao Tome & Principe (EMAE) 477 Source: World Bank (2025c) 40 41